Radiology Ontology for AI Datasets, Models and Projects
The Radiology Ontology for Artificial Intelligence Models, Datasets, and Projects (ROADMAP) provides a formal description of the metadata to index the growing number of artificial intelligence (AI) models and datasets, especially in diagnostic radiology.
ROADMAP builds upon generalized "model cards" and "datasheets for datasets" by highlighting features specific to medical imaging and by referencing concepts from related ontologies, coding schemes, and common data elements. In accordance with the FAIR guiding principles, application of the ontology will allow AI resources to be more readily discoverable and reusable. Its application also is expected to improve the ability to match AI models with relevant datasets and to facilitate detection of potential biases in released AI models.
-
The ontology in Manchester OWL syntax (open with Protégé)
- View in WebProtege - (WARNING: May not be current!)
-
The Metrics components of the ontology provides a compendium of AI model performance measures
- View an alphabetical list of the 191 scalar metrics
- View a list of metric preferred and alternate names